Backtest any trading strategy with AI — from plain English to results in minutes
Describe the idea. Labs writes the strategy code, runs the backtest, and returns the report. No coding, no data plumbing, no black box.
Most people with a strategy idea never test it. The idea dies in a spreadsheet: getting historical data, cleaning it, writing indicator code, sizing positions, simulating costs — it is a weekend of plumbing for every single idea.
Labs removes the plumbing. You type the strategy — "buy NVDA when RSI drops below 30 and the 50-day is above the 200-day, stop at 8%" — and the AI turns it into a strategy definition, then Labs runs it through a standardized pipeline: data ingestion, signal generation, position sizing, risk controls, and execution simulation.
What you get back
The report reads like a quant's, because it uses a quant's metrics — every run timestamped against the same dated data the public research pages cite:
- Sharpe ratio — return per unit of risk, so you can compare ideas fairly
- Max drawdown — the worst losing stretch you would have sat through
- Win rate and trade list — every simulated entry and exit, inspectable
- Equity curve — the shape of the journey, not just the destination
If you code, the generated Python (pandas, numpy, TA-Lib-style indicators) is fully inspectable and editable — nothing is hidden. Write once, run against history, then deploy to paper trading.
See it live before signing up
Every Labs run is ranked publicly by realized performance — timestamped and verifiable, no cherry-picked backtests, no survivorship bias. Browse real results first:
View the public Labs ranking → · See the live market dashboard →
Do I need to code?
No — describe the strategy in plain language and the AI writes the strategy definition. If you do code, you can inspect and edit everything it generates.
What does it cost?
The free plan includes 30 monthly AI credits — no card required. Each backtest shows its credit cost before you run it, and top-up packs from $12 never expire while your account is active.
Is a good backtest a promise of future returns?
No — a backtest is evidence about a process on past data, not a prophecy. That is exactly why every run is published publicly and timestamped, so nothing can be cherry-picked. Research, not advice.
Run your first backtest free
30 monthly credits included — no card required. Your strategy idea takes about 5 minutes to test.
← All guides · Home · Pricing